Struvite Control without Chemicals

-- A Case Study --

Michigan Water Environment Association

Biosolids/IPP Joint Conference

September 28, 2011

Jerald O. Thaler, P.E.

Fishbeck, Thompson, Carr & Huber, Inc.

1

Project Description

Michigan POTW

• 5 mgd design capacity

• RBC secondary treatment

• Chemical phosphorus removal

• Anaerobic digestion

Concrete-like deposits in supernatant pump

discharge piping of anaerobic digester

Evaluate and recommend control method

2

Agenda

The Struvite Problem

•Chemical control

•Nonchemical control

Water Chemistry Background

•Modeling pH

•Precipitation potential

Study Results and Discussion

3

The Struvite Problem

Digester-related deposits are typically

magnesium ammonium phosphate (struvite)• Reduce flow capacity, foul pumps, and damage valves

• Occur after point of turbulence or pressure drop

Emerged at WWTPs

during 1960s in parallel

with more widespread

phosphorus treatment

4

The Struvite Problem (cont’d)

May be more common with trend toward

Biological Phosphorus Removal (BPR)

•Phosphorus accumulating organisms

•More phosphorus release during anaerobic

digestion than with chemical precipitation

•Elevated levels of soluble phosphorus in digester

supernatant streams

Struvite formation can be unanticipated

consequence of switch to BPR

5

Control Methods

Prevent formation by sufficiently reducing

reactant ion(s)

Mg+2 + NH4+ + PO4

-3 + 6 H2O → MgNH4PO4•6H2O

Minimize build-up by eliminating turbulence

and/or using smooth pipe materials

Prevent formation for effectiveness over

long-term

6

Chemical Prevention

Reduce PO4-3 with metallic salts

• Ferrous/ferric chloride and alum most common

• Requires large chemical dose to be effective

• Increases inorganic fraction in biosolids

• Increases risk of forming other deposits

Ferrous phosphate (vivianite)

Aluminum silicates

Proprietary additives• Questionable effectiveness

• Expensive

7

Chemical Prevention (cont’d)

Lower pH with acid • Large dose required for any significant change

• Increases risk of corrosion

• Requires handling of hazardous material

Other• CO2 injection

• Magnesium hydroxide

.

.

.

8

Non-Chemical Prevention

Struvite control by dilution proposed in 1972• Borgerding Joseph, Phosphate Deposits in

Digestion Systems, Journal WPCF, 44, 813-819

Involves simultaneously reducing all three

reactant ions

Effectiveness varies; key is accounting for

site-specific water chemistry

9

Water Chemistry Background

pH is the master variable

Sets distribution of acid/base equilibria

NH4+ ↔ NH3

H3PO4 ↔ H2PO4- ↔ HPO4

-2 ↔ PO4-3

Easy to measure, but historically has been

difficult to calculate

pH calculations now more straightforward

using available software tools

10

pH Model Basics

In general, pH set by relative concentrations

of carbonate and hydroxide components

Primary carbonate components:• Dissolved carbon dioxide & carbonic acid (H2CO3*)

• Bicarbonate (HCO3-)

• Carbonate (CO3-2)

Primary hydroxide components:• Hydronium (H+)

• Hydroxide (OH-)

11

pH Model Basics

Three macro-variables for closed system

•pH (s.u.)

pH = -log H+

•Alkalinity (mg/L CaCO3)

ALK = HCO3- + 2*CO3

-2 + OH- – H+

•Total Inorganic Carbon (mg/L C)

TIC = H2CO3* + HCO3- + CO3

-2

12

pH Model Basics (cont’d)

pH, ALK, and TIC are complementary

• If know any two, mathematical solution exists

for the third

ALK and TIC are conservative

•Follow principal of mass balances

Key properties for predicting effect of

chemical addition, stream mixing, etc. e.g. Increase ALK with constant TIC → Higher pH

Decrease TIC with constant ALK → Higher pH

13

pH Model Examples

Chemical addition

• Stream 1: pH=7.5, ALK=125 → TIC=32

• Add 50 ppm soda ash (Na2CO3)

∆ALK=+47 and ∆TIC=+6

• Stream 1’: ALK=172, TIC=38 → pH=9.3

Stream Mixing

• Blend 50 gpm Stream 1’ with 150 gpm Stream 2

• Stream 2: pH=7.0, ALK=25 → TIC=7

• Mixed Stream: ALK=62, TIC=15 → pH=8.1

14

Precipitation Potential

Saturation index (Ω) is ratio of ion activity

product (IAP) to solubility constant (Ks):

Higher Ω indicates higher precipitation potential

•Deposit build-up typically starts at Ω = 5-10

•Controlled by rates of crystal nucleation and growth

Struvite reactant ions are all pH-dependent, so

Ω is also pH-dependent

15

pH Effect on Reactant Ions

16

pH Effect on Ω

0

100

200

300

400

500

7 8 9 10 11 12

Stru

vite

Sa

tura

tio

n I

nd

ex

pH, s.u.

Total P = 165 mg/LTotal Mg = 22 mg/L

Ammonia-N = 935 mg/L

17

pH Effect on Ω (cont’d)

18

Study Overview

Procedure

•Sampled for basic water chemistry parameters

•Measured pH upstream and downstream of

digester supernatant transfer pump

•Simulated effect of dilution, using pH model

and two alternate water sources:

– Well water

– WWTP effluent

19

Sampling Data

Basic water chemistry parameters:

Supernatant Transfer Pump

•Average ∆pH ~0.05 s.u.

•Calculated CO2 loss of 5-10% (via pH model)

Parameter Units Value

pH s.u. 7.85. ALK mg/L CaCO3 3,990 TIC mg/L C (__)* Dissolved Mg mg/L Mg 22 Total Phosphorus mg/L P 165 Total Ammonia mg/L N 935

*Calculated value

Parameter Digester

Supernatant

Well

Water

WWTP

Effluent

pH, s.u. 7.85 7.39 7.16

ALK, mg/L CaCO3 3,990 253 125

TIC, mg/L C (981)* (66)* (34)*

Temperature, oC 22 16 20

Dissolved Mg, mg/L 22 23 21

Total P, mg/L 165 0.007 0.68

Ammonia-N, mg/L 935 0.033 0.27

*Calculated (via pH model)

20

Simulation Basis

21

Dilution Effect on Mg+2

0

10

20

30

40

50

0% 25% 50% 75%

Mg

+2

,mg

/L

Dilution Ratio

Effluent

Well Water

22

Dilution Effect on NH4+

0

200

400

600

800

1000

0% 25% 50% 75%

NH

4+, m

g/L

Dilution Ratio

Effluent

Well Water

23

Dilution Effect on PO4-3

0.000

0.005

0.010

0.015

0.020

0.025

0% 25% 50% 75%

PO

4-3

, mg

/L

Dilution Ratio

Effluent

Wellwater

24

Dilution Effect on pH

7.5

7.6

7.7

7.8

7.9

8.0

0% 25% 50% 75%

pH

Dilution Ratio

Effluent

Wellwater

25

Site-Specific Control Curve

0

5

10

15

20

0% 25% 50% 75%

Stru

vite

Sat

ura

ito

n I

nd

ex

Dilution Ratio

Phase I (current)

Initial

Well Water

Effluent

Phase II (future)

26

Feedback from The Field

Six months into Phase I

“ I can say that it has helped. We haven’t had

any Moyno pumps quit from build-up on the

rotor.

Also, just past the pump is a reduced section of

pipe where the flow meter is located -- this was

the first place we found plugged with struvite.

No plugging has happened yet. “

27

Perspective

Struvite is a problem at some WWTPs

Expect to be more common with trend to BPR

Dilution is feasible control method, but water

chemistry must be thoroughly evaluated

Effective dilution ratios are

highly site-specific

May be possible to reuse

effluent for dilution water

28

Questions and Discussion

Thank You!

Jerald O. Thaler, P.E.Fishbeck, Thompson, Carr & Huber, Inc.

Lansing, MI517-887-4080 / jothaler@ftch.com

For further information, contact:

29